7/27/2019 Keita - Afroasiatic

http://slidepdf.com/reader/full/keita-afroasiatic 1/33 DECEMBER 2004 VOL 306 SCIENCE www.sciencemag.org680

Prescribed Fire and

Natural Disturbance

THE RECENT COVERAGE OF OUR WORK ON THE

relationship between fire history and an

emerging forest epidemic called suddenoak death highlights landscape-levelaspects of disease spread, which are oftenoverlooked (“Fighting sudden oak deathwith fire?”, J. Withgott, News Focus, 20Aug., p. 1101). Although we are interested in the possible role of prescribed fire inmanaging this disease, subsequent reportsin the popular press have claimed that weadvocate such an approach as treatment. Acautionary note is therefore required at this point. We have not found a direct connec-tion between fire suppression and thisdisease, and there is reason to suspect thatthe effects of past wildfires could be verydifferent than those of the typicalcontrolled burn. The decision to use prescribed fire in an ecosystem should beguided by location- and case-specificconsiderations (1).

As Lindenmayer  et al. note in their 

Policy Forum “Salvage harvesting policiesafter natural disturbance” (27 Feb., p.1303), natural disturbances such as fire areintegral to the healthy functioning of mostecosystems and are often poorly under-stood in policy and management arenas.The emphasis here is on “natural” distur- bances and the important role they play.Most prescribed burns, however, areattempted during conditions when fire isnot likely to escape control (e.g., outsidethe normal fire season). Burning under 

these conditions will not necessarily produce the natural range of f ire severitiesand subsequent fire effects that could result from past wildfires.

Restoring fire regimes is of greatimportance, but prescribed fires must ulti-

mately mimic natural events to fulfill their role in disturbance-mediated ecosystems.Prescribed fires that do not attain this goalcan have harmful ecological effects, even if successful for goals of fuel reduction and fire reintroduction. Populations of fire-dependent native species can be decimated (2) if timing or heating requirements for regeneration are not met. Invasive speciesmay also be promoted, which can lead tonear-permanent alteration of fire regimesand ecosystem functioning (3). Whether for ecosystem health in general, or management of forest pathogens in partic-ular, prescribed fire will need to be tailored to the societal goals and ecological require-ments of the situation at hand.

MAX A. MORITZ1 AND DENNIS C. ODION2

1Ecosystem Sciences Division, Department of 

Environmental Science, Policy, and Management,

University of California at Berkeley, Berkeley, CA

94720, USA. 2Institute for Computational Earth

Systems Science, University of California at Santa

Barbara, Santa Barbara, CA 93106, USA.

References1. S. Pyne, Science 294, 1005 (2001).2. D.Odion, C. Tyler, Conserv. Ecol. 6, 4 (2002).3. M.Brooks et al., BioScience 54, 677 (2004).

The Origins of 

Afroasiatic

IN THEIR REVIEW “FARMERS AND THEIR

languages: the first expansions” (25 Apr.2003, p. 597), J. Diamond and P. Bellwood suggest that food production and theAfroasiatic language family were broughtsimultaneously from the Near East toAfrica by demic diffusion, in other words, by a migration of food-producing peoples.In resurrecting this generally abandoned view, the authors misrepresent the views of 

the late I. M. Diakonoff (1), rely onlinguistic reconstructions inapplicable totheir claims (2), and fail to engage the fivedecades of Afroasiatic scholarship thatrebutted this idea in the first place. Thisextensive, well-grounded linguistic research places the Afroasiatic homeland in thesoutheastern Sahara or adjacent Horn of Africa (3–8) and, when all of Afroasiatic’s branches are included, strongly indicates a pre–food-producing proto-Afroasiaticeconomy (1, 7, 8).

A careful reading of Diakonoff (1shows his continuing adherence to hlong-held position of an exclusivelAfrican origin (4, 5) for the family. Hexplicitly describes proto-Afroasiatvocabulary as consistent with non – food producing vocabulary and links it to pre Neolithic cultures in the Levant and iAfrica south of Egypt, noting the latter t be older. Diakonoff does revise his location for the Common Semitic homelandmoving it from entirely within northeaAfrica to areas straddling the Nile Deltand Sinai, but continues to place thorigins of the five other branches of thAfroasiatic language family wholly iAfrica (1). One interpretation of tharchaeological data supports a pre–food producing population movement fromAfrica into the Levant (9), consistent witthe linguistic arguments for a pre-Neolithmigration of pre–proto-Semitic speakeout of Africa via Sinai (8).

The proto-language of each Afroasiat branch developed its own distinct vocabulary of food production, further supportinthe view that herding and cultivatioemerged separately in each branch after th proto-Afroasiatic period (7 , 8). Diamonand Bellwood adopt Militarev’s (2) solitarcounterclaim of proto-Afroasiatic cultivation. However, not one of Militarev proposed 32 agricultural roots can bconsidered diagnostic of cultivationFifteen are reconstructed as names o plants or loose categories of plants. Sucevidence may reveal plants known to earlAfroasiatic speakers, but it does not indcate whether they were cultivated or wildMilitarev’s remaining roots are eacsemantically mixed, i.e., they have food

 production–related meanings in somlanguages, but in other languages havmeanings applicable to foraging or equallapplicable to foraging or cultivating.

Furthermore, the archaeology onorthern Africa does not support demidiffusion of farming populations from th Near East. The evidence presented bWetterstrom (10) indicates that earlAfrican farmers in the Fayum initiallincorporated Near Eastern domesticateinto an indigenous foraging strategy, an

Skeletons of federally listed (threatened)Morro Manzanita shrubs ( Arctostaphylos morroensis) immediately after a prescribedburn, which led to its local extirpation (2).

Letters to the Editor Letters (~300 words) discuss material publishedin  Science in the previous 6 months or issuesof general interest. They can be submittedthrough the Web (www.submit2science.org)or by regular mail (1200 New York Ave., NW,Washington, DC 20005, USA). Letters are notacknowledged upon receipt, nor are authorsgenerally consulted before publication.Whether published in full or in part, letters aresubject to editing for clarity and space.

LETTERS

Published by AAAS

7/27/2019 Keita - Afroasiatic

http://slidepdf.com/reader/full/keita-afroasiatic 2/3

L E T T E R

www.sciencemag.org SCIENCE VOL 306 3 DECEMBER 2004

only over time developed a dependence onhorticulture. This is inconsistent with in-migrating farming settlers, who would have brought a more abrupt change insubsistence strategy. The same archaeolog-ical pattern occurs west of Egypt, wheredomestic animals and, later, grains weregradually adopted after 8000 yr B.P. intothe established pre-agricultural Capsianculture, present across the northern Sahara

since 10,000 yr B.P. (11

). From this conti-nuity, it has been argued that the pre–food- production Capsian peoples spoke languagesancestral to the Berber and/or Chadic branches of Afroasiatic, placing the proto-Afroasiatic period distinctly before 10,000yr B.P. (8). Furthermore, there is evidencethat cattle domestication occurred inde- pendently in the early Holocene easternSahara, earlier than in the Near East (12),casting doubt on the idea of a single originof food production in the Levant.

A critical reading of genetic dataanalyses, specifically those of Y chromo-some phylogeography and TaqI 49a,f haplotypes, supports the hypothesis of  populations moving from the Horn or southeastern Sahara northward to the NileValley, northwest Africa, the Levant, and Aegean (13–15). The geography of theM35/215 (or 215/M35) lineage, which is of Horn/East African origin, is largelyconcordant with the range of Afroasiaticlanguages. Underhill et al. state that thislineage was carried from Africa during the“Mesolithic” (13). The distributions of theAfroasiatic branches and this lineage can best be explained by invoking movementsthat originated in Africa and occurred  before the emergence of food production,as well as after.

CHRISTOPHER EHRET,1 S.O.Y. KEITA,2

PAUL NEWMAN3

1Department of History,University of California at

Los Angeles, Los Angeles, CA 90095, USA.2National Human Genome Center at Howard

University, Howard University Hospital,

Washington, DC 20060, USA, and Department of 

Anthropology, Smithsonian Institution, National

Museum of Natural History, Washington, DC

20560, USA. 3Department of Linguistics, Indiana

University, Bloomington, IN 47405, USA.

References1. I. M. Diakonoff, J. Semit. Stud. 43, 209 (1998).2. A. Militarev, in Examining the Farming/Language

Dispersal Hypothesis, P. Bellwood, C. Renfrew, Eds.(McDonald Institute for Archaeological Research,Cambridge, 2003), chap. 12.

3. J. H. Greenberg,  Studie s in African Linguis ticClassification (Compass Publishing, New Haven, CT,1955).

4. I. M. Diakonoff, Altorientalische Forschung. 8, 23 (1981).5. I. M. Diakonoff,  Afrasian Languages (Nauka Publishing

House, Moscow, 1988).6. H. L. Fleming, in The Non-Semitic Languages of 

Ethiopia, M. L. Bender, Ed. (Michigan State University,African Studies Center, East Lansing, MI, 1976), pp.298–323.

7. C. Ehret, J. Afr. Hist. 20, 161 (1979).8. C. Ehret,in Symposium 13d:Rock Art and the Sahara, in

Proceedings of the International Rock Art and Cognitive Archaeology Congress, A. Muzzolini, J.-L. Le Quellec,Eds. (Centro Studie Museo d’Arte Prehistorica, Turin,Italy, 1999) (HTML-CD Rom edition, ehlist1.jpg).

9. O. Bar Yosef, Afr. Archaeol. Rev . 5, 29 (1987).10. W. Wetterstrom, in  Archaeology of Africa, T. Shaw et 

al., Eds. (Routledge, London, 1993), pp. 165–226.11. N. Rahmani, Le Capsien typique et le Capsien

 supérieur , Cambridge Monographs in Archaeology 57(Cambridge Univ. Press, Cambridge, 2003).

12. F. Wendorf  et al., Eds., Holocene Settlement of theEgyptian Sahara, vol. 1, The Archaeology of NabtaPlaya (Kluwer Academic/Plenum Publishers, New York,2001).

13. P. Underhill et al., Am. J. Hum. Genet . 65, 43 (2001).14. G. Lucotte, G. Mercier, Am. J. Phys. Anthropol. 121, 63

(2003).15. O. Semino et al., Am. J. Hum. Genet. 74, 1023 (2004).

ResponseEHRET E T AL. SUGGEST THAT EARLY

Afroasiatic languages were spread byMesolithic foragers from Africa into theLevant. In our Review, we did not positivelyfavor either the African or the Levant originhypothesis (p. 601). But in the map (Fig. 2), I

chose the Levant hypothesis, because I believe, on balance, that it provides the bestexplanation for the evidence that has survived through 12,000 years of prehistory.

In linguistic terms, Ehret (1) has presented a phylogenetic history for Afroasiaticlanguages, based on shared phonologicalinnovations, that contains a primary division between the Omotic languages of Ethiopiaand an Erythraean subgroup that includes allother Afroasiatic languages (includingSemitic and Ancient Egyptian). This ordering,if correct, suggests an African origin for thefamily. But is it correct? Diakonoff (2, 3) hasoffered a completely different grammaticalsubgrouping structure for Afroasiatic, in the process, casting doubt on Omotic as amember of the family and suggesting [(2), p.218] that the predomestication [but probablyearly cultivating (4)] Natufian archaeologicalcomplex of Palestine matches well with proto-Afrasian (Afroasiatic) cultural and environmental vocabulary reconstructions.Militarev’s reconstructed proto-Afroasiaticvocabulary (5), whether “agricultural” or not,is also peopled with animals and plants of Levant, not African, origin and matches a Natufian cultural landscape. Ehret et al. pointout that Militarev’s semantic reflexes are

mixed, but perhaps this is to be expected giventhat plants of Levant (winter rainfall) origindid not spread prehistorically into the desertor summer rainfall belts of northern Africa beyond the Mediterranean coast, Egypt, and highland Ethiopia.

In archaeological terms, I agree thatearly Saharans managed cattle, and Ehrethimself convincingly relates the earliestappearance of this tradition to Nilo-Saharan–speaking populations (6 ). TheEgyptian Neolithic economy, however, was

manifestly of Levant and not Africorigin. Domesticated sheep and goats we probably introduced via Arabia into tHorn of Africa at a similar time, circa sixmillennium B.C.

My assumption is that the spread Afroasiatic occurred as a result of actuhuman movement, not language diffusialone. There is no significant archaeoloical evidence for a population moveme

from Africa into the Levant, whethMesolithic or Neolithic, at the time question. The genetics papers quoted bEhret et al. do not settle this matter. Thechromosome evidence appears to signcomplex two-way population movemenwith very uncertain chronologies. Mworking assumption, therefore, is thearly Afroasiatic languages spread frothe Levant into Africa between 7000 a12,000 years ago, probably in more thone movement. Subsequent history hseen an enormous spread of Semitlanguages, including Ethiopian Semitand, of course, Arabic, on such a scale ththe original phylogenetic geography of tAfroasiatic language family must ha been considerably erased. Because of ththe geographical source of this family wnot reveal itself easily. I have ju published a detailed discussion Afroasiatic prehistory from archaeologicand linguistic perspectives (4), and tabove points are made in more detail ther

PETER BELLWO

Department of Archaeology & Anthropolog

Australian National University, Canberra, A

0200, Australia.

References1. C. Ehret, Reconstructing Proto-Afro-Asiatic (Univ

California Press, Berkeley, CA, 1995).2. I. M. Diakonoff, J. Semit. Stud. 43, 209 (1998).3. I. M. Diakonoff, J. Near Eastern Stud. 55, 293 (19964. P. Bellwood, First Farmers (Blackwell, Oxford, 200

pp. 97–106, 207–210.5. A. Militarev, in Examining the Farming/Langua

Dispersal Hypothesis, P. Bellwood, C. Renfrew, E(McDonald Institute for Archaeological ResearCambridge, 2003), chap. 12.

6. C. Ehret, in Examining the Farming/Language DisperHypothesis, P. Bellwood, C. Renfrew, Eds. (McDonInstitute for Archaeological Research, Cambrid2003), chap. 14.

Earth’s EntropyRALPH LORENZ’S PERSPECTIVE “FULL STEA

ahead—probably” (7 Feb. 2003, p. 837) the recent groundbreaking work Roderick Dewar (1) mentions the puzzthat “All else being equal, MEP [maximuentropy production] would predict  planet’s meridional temperature contrast  be independent of its rotation rate. Thdisagrees with some rudimentary GC[general circulation model] experimenand with meteorologists’ intuition.”

Published by AAAS

7/27/2019 Keita - Afroasiatic

http://slidepdf.com/reader/full/keita-afroasiatic 3/3

It is well known that tidal and atmos- pheric motions exert torque on the solid Earth, which detectably affects its rotationrate (2, 3). Hadley-cell–driven tradewinds, for example, exert torque onEarth’s surface in a direction that promotes continued rotation. This could conceivably amount to ordered work thatacts as an additional mode of entropy production. Perhaps climate modelers

should investigate whether one conse-quence of maximum entropy productionon Earth may be partial regulation of plan-etary rotation rate.

NATHAN G. PHILLIPS

Geography Department, Boston University, 675

Commonwealth Avenue, Boston, MA 02215, USA.

References1. R.L. Dewar, J.Phys. A.Math.Gen. 36, 631 (2003).2. See http://badc.nerc.ac.uk/data/aam/.3. See http://tycho.usno.navy.mil/leapsec.990505.html.

Response

PHILLIPS SUGGESTS THAT THERMODYNAMICS

may guide planetary rotations. For Earth,at least, this is unlikely to be so. Theusefulness of maximum entropy produc-tion (MEP) is only as a selection guidelineamong dynamically permitted steadystates, and the rotation state of the planetmay control which states are dynamically possible. The system must first complywith the rigid laws of physics, notably theconservation of mass, energy, and angular momentum: These factors are imposed asconstraints on the system before MEPapplies.

Even if Earth’s whole atmosphere wereto spin up to the speed of sound (an extremecase!), angular momentum balance meansthe rotation period of the solid Earth (wheremuch of the solar heat is absorbed and reradiated) changes by only about one partin one million—a level unlikely to affectheat transfer. Thus, even if the dynamicsallowed such a spin-up, it seems thesystem would gain little from the effort.

However, Phillips’ basic suggestion,that optimality in heat transport may guiderotation rates, may have merit for theatmospheres of extrasolar giant planets (1)where atmospheric motions at the rela-

tively high altitudes where starlight isabsorbed and thermal radiation emitted are largely decoupled from the motion of the planet’s interior. If the motions areguided by an MEP heat transport criterion,close-in extrasolar planets, even if tidallylocked to their parent star, may nonethe-less have only modest day:night tempera-ture contrasts.

RALPH D. LORENZ

Lunar and Planetary Laboratory, University of 

Arizona,Tucson,AZ 85721, USA.

Reference1. J. I. Lunine, R. D. Lorenz, “A simple prescription for 

calculating day-night temperature contrasts onsynchronously rotating planets,” 33rd Annual Lunar and Planetary Science Conference, 11 to 15 March2002, Houston, TX, abstr. no.1429.

The Brain, Neurons, and

Behavior

I OPENED THE SPECIAL ISSUE ON COGNITION

and Behavior (15 Oct., pp. 431–452) witha “there we go again” feeling. So it was arelief to read Donald Kennedy’s Editorial“Neuroscience and neuroethics” (p. 373).It has become fashionable to equate the brain with the mind, which in turn controls behavior. Presumably it’s hard science, because neurons are involved. But it isn’t.It’s just a confusion of the necessary withthe sufficient, a point made in theEditorial. Altogether too often, sight is lostof the fact that any particular brain canevolve into any particular mind, dependingon the experiences encountered.

 JOSEPH M. NOTTERMAN

Department of Psychology, Princeton University,

Princeton, NJ 08544, USA.

CORRECTIONS AND CLARIFICATIONS

News Focus: “RNAi shows cracks in its armor” by J. Couzin (12 Nov., p. 1124). On page 1125, in thesecond column, second paragraph, the sentence,“At a meeting last week in Titisee, Germany, Sharppresented preliminary data from his lab showing a

10-fold change in protein levels with only atwofold microRNA difference, the level commonlyseen from an off-target effect,” the term“microRNA” should have read “mRNA.”

Random Samples: “Good as new”(5 Nov., p. 971).This item incorrectly reported that a new laser technique for cleaning ancient coins was devel-oped by Italian archaeologists. It was devised byphysicists at IFAC-CNR in Florence, Italy. Theaccompanying photo credit should have read S.Siano.

Reports: “Requirement for caspase-2 in stress-induced apoptosis before mitochondrial perme-

abilization” by P. Lassus et al. (23 Aug. 2002, p.1352). This paper reported that silencing expres-sion of caspase-2 with an siRNA prevented apop-tosis. Since the time of publication, the authorshave identified an siRNA that silences expressionof the caspase-2 protein but fails to prevent apop-tosis.The authors are investigating three possibili-ties to explain their results: (i) These siRNAs differ-entially silence caspase-2 isoforms, which altersthe outcome of drug-induced apoptosis; (ii) one of the two siRNAs silences an unidentified gene(s),whose product is involved in apoptosis; and (iii)one of the two siRNAs has some effect unrelatedto RNAi.

L E T T E R S

www.sciencemag.org SCIENCE VOL 306 3 DECEMBER 2004PublishedbyAAAS